tensorflow 问题汇总，持续更新...

最新推荐文章于 2022-03-26 11:41:51 发布

HelloJFS

最新推荐文章于 2022-03-26 11:41:51 发布

阅读量568

点赞数

分类专栏： tensorflow

本文链接：https://blog.csdn.net/HelloJFS/article/details/82021110

版权

6 篇文章 0 订阅

订阅专栏

（1） TypeError: Expected int32, got list containing Tensors of type '_Message' instead.
（2） TypeError: Expected int32, got <tf.Variable 'PolicyNetwork/pad_embed:0' shape=(1, 200) dtype=float32_ref> of type 'Variable' instead.

解决方法：

查了很多都说要重新装TensorFlow，然后找到了下面的解决方法：

You just have to update all the line with tf.concat

for example

net = tf.concat(3, [branch_0, branch_1, branch_2, branch_3])

should be changed to

net = tf.concat([branch_0, branch_1, branch_2, branch_3], 3)

tf.mul, tf.sub and tf.neg are deprecated in favor of tf.multiply, tf.subtract and tf.negative.

You'll need to replace tf.mul with tf.multiply.

（4） TypeError: Input 'split_dim' of 'Split' Op has type float32 that does not match expected type of int32.

The documentation clearly indicates the calling order for tf.split should be:

tf.split(split_dim, num_split, value, name='split')

But the actual code indicates:

def split(value, num_or_size_splits, axis=0, num=None, name="split"):

（5）AttributeError: module 'tensorflow.tools.api.generator.api.nn' has no attribute 'rnn'

在新版本中，tf.nn.rnn 被 tf.nn.static_rnn 替换。

（6）ValueError: Only call `softmax_cross_entropy_with_logits` with named arguments (labels=..., logits=..., ...)

tf.nn.softmax_cross_entropy_with_logits(logits_expanded,labels)

更换为：

tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits_expanded,labels=labels))

But instead of calling g.__next__(), as Paolo says, use next(g).

关注